The influence of task context on neural population activity during natural scene discrimination

项目摘要/摘要 环境背景严重影响我们的视觉感知，使我们能够提取相关信息 来自自然界中发现的压倒性的感官输入。在以下情况下，这一点尤为重要 我们需要区分自然环境中的感觉输入，比如在人群中寻找朋友。语境 可以分为两个因素，可能会影响我们的辨别能力：我们有多熟悉 视觉输入，以及我们所期望的输入。调查语境影响的研究 调制主要使用任务中的几个简化刺激来评估单个神经元的活动 这两个成分是分不开的。这极大地限制了我们将这些发现推广到 熟悉和期望影响神经种群活动并塑造我们的 感知能力。这项建议的首要目标是调查任务情境对塑造的影响 神经群体反应和行为表现。在第一个具体目标中，我们将评估 调节刺激熟悉度对神经活动和行为的影响。我们假设刺激熟悉度的增加 通过调节群体活动来增强刺激的编码，从而提高我们的感知能力。在……里面 第二个具体目标，我们将评估调节对刺激将如何变化的预期的效果 神经活动和行为。我们假设，预期会改变人口活动，从而使 当前任务背景和提高行为绩效。我们采用人口水平措施的战略将 揭示只测量单个神经元的分析隐藏的信号。这项研究的结果将确立 神经种群活动如何根据环境的不同方面进行调整以改善我们的视觉 自然主义环境中的感知。这将增强我们对基本神经机制的理解 在非疾病状态下影响视觉感知，使开发治疗方法成为可能 视知觉失调的疾病。 除了一个令人兴奋和创新的研究项目外，这项提案还概述了一个全面的培训 这项计划将为我提供必要的培训，以迈向独立研究人员的职业生涯。这 PLAN强调1)对视觉感知的神经基础有更深的概念性理解， 2)学习必要的实验技术和计算分析 人口活动转化为行为，3)提高科学交流技能，4)拓宽我的科学网络， 以及5)为我提供指导他人的机会。培训将在以下人员的指导下进行 马修·史密斯博士是一位将神经群体活动与视觉感知联系起来的专家 匹兹堡大学和卡内基梅隆大学共同提供了一个杰出的神经科学社区 在横跨神经科学各个领域的世界知名教职员工中，广泛的跨学科研究和 合作，以及在培训未来独立调查人员方面的令人印象深刻的记录。

PROJECT SUMMARY/ABSTRACT Environmental context heavily influences our visual perception, allowing us to extract relevant information from the overwhelming amount of sensory input found in the natural world. This is particularly important when we need to discriminate sensory inputs in our natural environment, such as finding a friend in a crowd. Context can be separated into two components that could affect our ability to make discriminations: how familiar we are with the visual input, and what we expect the input to be. Studies investigating the effects of contextual modulation have primarily assessed the activity of individual neurons using only a few simplified stimuli in tasks where these two components can’t be separated. This greatly limits our ability to generalize these findings to natural environments where familiarity and expectation influence neural population activity and shape our perceptual abilities. The overarching goal of this proposal is to investigate the effects of task context on shaping neural population responses and behavioral performance. In the first specific aim, we will assess the effect of modulating stimulus familiarity on neural activity and behavior. We hypothesize that increased stimulus familiarity improves our perceptual abilities by modulating population activity to enhance the encoding of the stimulus. In the second specific aim, we will assess the effect of modulating the expectation of how a stimulus will change on neural activity and behavior. We hypothesize that expectation modifies population activity to benefit the current task context and improve behavioral performance. Our strategy to employ population-level measures will reveal signals hidden from analyses that only measure single neurons. The results of this study will establish how neural population activity is tailored to different aspects of an environmental context to improve our visual perception in a naturalistic environment. This will enhance our understanding of the basic neural mechanisms affecting visual perception in a non-disease state, making it possible to progress in developing treatments for disorders where visual perception is awry. In addition to an exciting and innovative research project, this proposal outlines a comprehensive training plan that will provide me with essential training to advance towards a career as an independent researcher. This plan emphasizes 1) gaining a deeper conceptual understanding of the neural underpinnings of visual perception, 2) learning the necessary experimental techniques and computational analyses required to relate neuronal population activity to behavior, 3) improving scientific communication skills, 4) broadening my scientific network, and 5) providing me with opportunities to mentor others. The training will be conducted under the guidance of Dr. Matthew Smith, an expert in relating neural population activity to visual perception, at the University of Pittsburgh and Carnegie Mellon University which together provide an outstanding neuroscience community rich in world-renowned faculty spanning every domain of neuroscience, extensive interdisciplinary research and collaboration, and an impressive track-record for training future independent investigators.